This work provides a geometric interpretation of FM uncertainty in the velocity field, showing that uncertainty manifests as deviation from an ideal affine-isotropic contraction field, and introduces denoising acceleration ( accel), a highly-generalizable and cost-free uncertainty proxy that measures the bending of the...
Flow-matching policies are promising for imitation learning because they model complex multimodal action distributions. However, their stochasticity is largely passive: repeated sampling may yield diverse behaviors, but users cannot directly choose among valid continuations from the same state. We propose Source-Lifted...
He Zhang, Ying Sun, Pengteng Li et al.· 0 citations
Self-patching identifies activation locations where relocating representations substantially improves failed generalization cases and design a simple heuristic strategy which recovers 58--75\% of the oracle headroom in generalization failure.
Five representative interfaces are implemented -- discrete state prompt, VLM prefix, action prefix, state expert, and feature modulation -- under matched implementation details, and evaluated on 45 atomic tasks spanning three task families plus 20 composite tasks.
Yiren Zhao, Ziyang Chen, Zi-Yang Rao et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
